U.S. patent application number 15/021891 was filed with the patent office on 2016-08-04 for monitoring camera, monitoring system, and motion detection method.
This patent application is currently assigned to MITSUBISHI ELECTRIC CORPORATION. The applicant listed for this patent is MITSUBISHI ELECTRIC CORPORATION. Invention is credited to Toshiaki SHIMADA.
Application Number | 20160225160 15/021891 |
Document ID | / |
Family ID | 52743587 |
Filed Date | 2016-08-04 |
United States Patent
Application |
20160225160 |
Kind Code |
A1 |
SHIMADA; Toshiaki |
August 4, 2016 |
MONITORING CAMERA, MONITORING SYSTEM, AND MOTION DETECTION
METHOD
Abstract
A monitoring camera including a picture shooter to perform
shooting on a target area to acquire a picture, a block partitioner
(a motion detecting unit) to partition the picture acquired by the
picture shooter into a plurality of blocks, a detection sensitivity
acquirer (an input output unit) to acquire detection sensitivity
for each of the blocks into which the picture is partitioned by the
block partitioner, a motion detector to determine the presence or
absence of a motion for each of the blocks, into which the picture
is partitioned by the block partitioner, according to the
corresponding detection sensitivity acquired by the detection
sensitivity acquirer, and an outputter (the input output unit) to
output a detection result acquired by the motion detector.
Inventors: |
SHIMADA; Toshiaki;
(Chiyoda-ku, JP) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
MITSUBISHI ELECTRIC CORPORATION |
Tokyo |
|
JP |
|
|
Assignee: |
MITSUBISHI ELECTRIC
CORPORATION
Chiyoda-ku, Tokyo
JP
|
Family ID: |
52743587 |
Appl. No.: |
15/021891 |
Filed: |
September 26, 2014 |
PCT Filed: |
September 26, 2014 |
PCT NO: |
PCT/JP2014/075714 |
371 Date: |
March 14, 2016 |
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
G01B 11/14 20130101;
G06T 7/60 20130101; G06T 2207/30244 20130101; G06T 7/70 20170101;
H04N 5/23218 20180801; G06T 7/20 20130101; G06T 2207/10016
20130101; G08B 13/19602 20130101; G06T 2207/30232 20130101; H04N
7/18 20130101; G06T 2207/20021 20130101; H04N 5/232 20130101; H04N
5/23206 20130101; G06T 7/50 20170101; G06T 7/254 20170101 |
International
Class: |
G06T 7/20 20060101
G06T007/20; G01B 11/14 20060101 G01B011/14; G06T 7/60 20060101
G06T007/60; H04N 7/18 20060101 H04N007/18; G06T 7/00 20060101
G06T007/00 |
Foreign Application Data
Date |
Code |
Application Number |
Sep 26, 2013 |
JP |
2013-199833 |
Claims
1: A monitoring camera comprising: a picture shooter to perform
shooting on a target area to acquire a picture; a block partitioner
to partition the picture acquired by said picture shooter into a
plurality of blocks; a detection sensitivity acquirer to acquire
detection sensitivity for each of the blocks into which the picture
is partitioned by said block partitioner; a motion detector to
determine presence or absence of a motion for each of the blocks,
into which the picture is partitioned by said block partitioner,
according to corresponding detection sensitivity acquired by said
detection sensitivity acquirer; and an outputter to output a
detection result acquired by said motion detector.
2: The monitoring camera according to claim 1, wherein said motion
detector determines whether or not a block having a motion
continuing for a set time period or longer is included in the
blocks into which the picture is partitioned by said block
partitioner.
3: The monitoring camera according to claim 2, wherein said
monitoring camera includes a set time period acquirer that acquires
information indicating said set time period, and said motion
detector uses the set time period acquired by said set time period
acquirer.
4: The monitoring camera according to claim 1, wherein said motion
detector determines whether or not a number of blocks each having a
motion, among the blocks into which the picture is partitioned by
said block partitioner, is equal to or greater than a set block
number.
5: The monitoring camera according to claim 4, wherein said
monitoring camera includes a set block number acquirer that
acquires information indicating said set block number, and said
motion detector uses the set block number acquired by said set
block number acquirer.
6: The monitoring camera according to claim 1, wherein said
monitoring camera includes a block information acquirer that
acquires information indicating a size of said block in each of a
plurality of regions into which said picture is divided, and said
block partitioner partitions said picture into blocks having a
plurality of sizes according to the size of the block acquired by
said block information acquirer.
7: The monitoring camera according to claim 6, wherein said
monitoring camera includes a distance acquirer that acquires
information indicating a distance between said target area in said
block and the monitoring camera itself, and said block information
acquirer acquires the size of said block from the distance acquired
by said distance acquirer.
8: The monitoring camera according to claim 5, wherein said
monitoring camera includes a distance acquirer that acquires
information indicating a distance between said target area in said
block and the monitoring camera itself, and said set block number
acquirer acquires said set block number from the distance acquired
by said distance acquirer.
9: The monitoring camera according to claim 1, wherein said
detection sensitivity acquirer acquires said detection sensitivity
set according to a distance between said target area in said block
and a reference point.
10: The monitoring camera according to claim 9, wherein said
reference point is an installation position of the monitoring
camera itself.
11: The monitoring camera according to claim 1, wherein said
monitoring camera includes a distance acquirer that acquires
information indicating a distance between said target area in said
block and a reference point, and said detection sensitivity
acquirer acquires said detection sensitivity from the distance
acquired by said distance acquirer.
12: The monitoring camera according to claim 11, wherein said
reference point is an installation position of the monitoring
camera itself.
13: The monitoring camera according to claim 1, wherein said
detection sensitivity acquirer acquires said detection sensitivity
set according to an influence of light upon said target area in
said block.
14: The monitoring camera according to claim 1, wherein said
monitoring camera includes an optical influence information
acquirer that acquires information indicating an influence of light
upon said target area in said block, and said detection sensitivity
acquirer acquires said detection sensitivity from the influence of
light acquired by said optical influence information acquirer.
15: The monitoring camera according to claim 1, wherein zoom
magnifying power for shooting is made to be variable, and said
detection sensitivity acquirer updates said detection sensitivity
according to said zoom magnifying power.
16: The monitoring camera according to claim 1, wherein said
detection sensitivity acquirer acquires said detection sensitivity
from a distance between said target area and the monitoring camera
itself which is measured by a sensor.
17: A monitoring system comprising: a picture shooter to perform
shooting on a target area to acquire a picture; a block partitioner
to partition the picture acquired by said picture shooter into a
plurality of blocks; a detection sensitivity acquirer to acquire
detection sensitivity for each of the blocks into which the picture
is partitioned by said block partitioner; a motion detector to
determine presence or absence of a motion for each of the blocks,
into which the picture is partitioned by said block partitioner,
according to corresponding detection sensitivity acquired by said
detection sensitivity acquirer; and an outputter to output a
detection result acquired by said motion detector.
18: A motion detection method comprising: a shooting step of a
picture shooter performing shooting on a target area to acquire a
picture; a block partitioning step of a block partitioner
partitioning the picture acquired by said picture shooter into a
plurality of blocks; a detection sensitivity acquiring step of a
detection sensitivity acquirer acquiring detection sensitivity for
each of the blocks into which the picture is partitioned by said
block partitioner; and a motion determining step of a motion
detector determining presence or absence of a motion for each of
the blocks, into which the picture is partitioned by said block
partitioner, according to corresponding detection sensitivity
acquired by said detection sensitivity acquirer.
Description
FIELD OF THE INVENTION
[0001] The present invention relates to a monitoring camera for, a
monitoring system for, and a motion detection method of performing
motion determination on the basis of a picture acquired by
performing shooting on a target area.
BACKGROUND OF THE INVENTION
[0002] Generally, when it is determined that an invader is staying
in a target area for monitoring, a monitoring system provides a
warning for outside the system. This determination of the presence
or absence of an invader is carried out typically by performing
motion determination on a picture acquired by performing shooting
on the target area (e.g., refer to patent reference 1).
[0003] In a conventional technology disclosed in the patent
reference 1, a target area is monitored by a plurality of monitor
stations. A monitoring camera partitions a picture acquired by
performing shooting on that target area into a plurality of blocks
and transmits a feature quantity acquired for each of the blocks to
the monitor stations. Each monitor station then performs motion
determination on the basis of the feature quantity received thereby
and acquired for each of the blocks by using a sensitivity
parameter for a detecting process which the monitor station
has.
RELATED ART DOCUMENT
Patent Reference
[0004] Patent reference 1: Japanese Unexamined Patent Application
Publication No. 2007-19759
SUMMARY OF THE INVENTION
Problems to be Solved by the Invention
[0005] However, in the conventional technology disclosed in the
patent reference 1, a single sensitivity parameter is used for each
picture. Therefore, there is a problem that objects are detected
locally with an excessive frequency and, conversely, objects are
hard to detect, according to the distance to the monitoring camera,
the sizes or patterns of objects, or a sporadic change of
illumination.
[0006] On the other hand, in conventional technologies, a plurality
of monitor stations are needed when motion determination is
performed by using a sensitivity parameter different for each
block. Therefore, there is a problem that the monitoring system
becomes complicated and the cost becomes high. Another problem is
that the number of blocks into which each picture is partitioned is
the order of the number of monitor stations.
[0007] Further, in conventional technologies, it is necessary to
transmit the feature quantity acquired for each block from the
monitoring camera to the monitor stations. Therefore, there is a
problem that a large-volume data transfer is needed between the
monitoring camera and the monitor stations. For example, in a
system or the like that needs to perform surveillance in real time,
it is necessary to secure a transmission capacity between the
monitoring camera and the monitor stations.
[0008] The present invention is made in order to solve the
above-mentioned problems, and it is therefore an object of the
present invention to provide a monitoring camera, a monitoring
system and a motion detection method capable of preventing
erroneous motion detection and omissions in motion detection in
each of blocks into which a picture is partitioned, with a simple
and low-cost configuration.
Means for Solving the Problem
[0009] In accordance with the present invention, there is provided
a monitoring camera including: a picture shooter to perform
shooting on a target area to acquire a picture; a block partitioner
to partition the picture acquired by the picture shooter into a
plurality of blocks; a detection sensitivity acquirer to acquire
detection sensitivity for each of the blocks into which the picture
is partitioned by the block partitioner; a motion detector to
determine the presence or absence of a motion for each of the
blocks, into which the picture is partitioned by the block
partitioner, according to the corresponding detection sensitivity
acquired by the detection sensitivity acquirer; and an outputter to
output a detection result acquired by the motion detector.
Advantages of the Invention
[0010] Because the monitoring camera in accordance with the present
invention is configured as above, the monitoring camera can prevent
erroneous motion detection and omissions in motion detection in
each of the blocks, into which the picture is partitioned, with a
simple and low-cost configuration.
BRIEF DESCRIPTION OF THE FIGURES
[0011] FIG. 1 is a block diagram showing the configuration of the
whole of a monitoring system including a monitoring camera in
accordance with Embodiment 1 of the present invention;
[0012] FIG. 2 is a block diagram showing the configuration of the
monitoring camera in accordance with Embodiment 1 of the present
invention;
[0013] FIG. 3 is a diagram showing picture partitioning for
determination of a motion by the monitoring camera in accordance
with Embodiment 1 of the present invention;
[0014] FIG. 4 is a diagram showing a relation between a motion
determination index value and a sensitivity parameter, the relation
being set according to variations in the distance from the
monitoring camera in accordance with Embodiment 1 of the present
invention to a target area, and FIG. 4(a) is a diagram showing a
case of using a single sensitivity parameter and FIG. 4(b) is a
diagram showing a case of using a plurality of sensitivity
parameters;
[0015] FIG. 5 is a diagram showing an example of a monitor picture
acquired by the monitoring camera in accordance with Embodiment 1
of the present invention;
[0016] FIG. 6 is a flow chart showing operations by the whole of
the monitoring system in accordance with Embodiment 1 of the
present invention;
[0017] FIG. 7 is a flow chart showing the operation of a motion
detector in Embodiment 1 of the present invention;
[0018] FIG. 8 is a diagram showing an example of a monitor picture
acquired by a monitoring camera in accordance with Embodiment 2 of
the present invention (when a person moves);
[0019] FIG. 9 is a diagram showing an example of determination of
the continuity of a motion which is performed by the monitoring
camera in accordance with Embodiment 2 of the present invention,
and FIG. 9(a) is a diagram showing a temporal change of the motion
determination index value which is caused by an influence of light
and FIG. 9(b) is a diagram showing a temporal change of the motion
determination index value which is caused by a motion of a moving
object;
[0020] FIG. 10 is a flow chart showing the operation of a motion
detector in Embodiment 2 of the present invention;
[0021] FIG. 11 is a diagram showing an example of a monitor picture
acquired by a monitoring camera in accordance with Embodiment 3 of
the present invention (when a person and a small animal are
staying);
[0022] FIG. 12 is a flow chart showing the operation of a motion
detector in Embodiment 3 of the present invention;
[0023] FIG. 13 is a diagram showing an example of a block
partitioning method for use in a monitoring camera in accordance
with Embodiment 4 of the present invention;
[0024] FIG. 14 is a flow chart showing the operation of a motion
detector in Embodiment 6 of the present invention;
[0025] FIG. 15 is a diagram explaining a relation between a
reference point and the distance from the reference point to a
target area in a monitoring camera in accordance with Embodiment 7
of the present invention, and FIG. 15(a) is a diagram showing a
case in which the reference point is the installation position of
the monitoring camera, FIG. 15(b) is a diagram showing a case in
which the reference point is a position which is the furthest from
the monitoring camera and FIG. 15(c) is a diagram showing a case in
which the reference point is a position between the installation
position of the monitoring camera and the position which is the
furthest from the monitoring camera; and
[0026] FIG. 16 is a diagram showing an example of a monitor picture
acquired by a monitoring camera in accordance with Embodiment 8 of
the present invention, and FIG. 16(a) is a diagram showing a
picture before enlargement and FIG. 16(b) is a diagram showing a
picture in which a part (a region C) of a zone A shown in FIG.
16(b) is enlarged.
EMBODIMENTS OF THE INVENTION
[0027] Hereafter, the preferred embodiments of the present
invention will be explained in detail with reference to the
drawings.
Embodiment 1
[0028] First, problems to be solved by Embodiment 1 will be
explained.
[0029] In the case of providing only one kind of sensitivity
parameter used for performing motion determination, like in the
case of a conventional configuration disclosed in patent reference
1, there may occur a phenomenon in which it is easily determined
(this determination is sometimes referred to as motion detection
from here on) that "there is a motion" in a certain region within
one picture while it is difficult to determine that "there is a
motion" in another region within the picture.
[0030] For example, there may occur a phenomenon in which if the
sensitivity parameter is set to a single certain value, when,
within a picture, a person is moving in a region closer to the
monitoring camera and another person is similarly moving in a
region farther from the monitoring camera, the motion of the person
who is closer to the monitoring camera is easily detected while the
motion of the other person who is farther from the monitoring
camera is hard to detect. This is because in the case of using a
method of performing motion determination according to the size of
a motion, a change state of the pattern of an object, a change
state of the pixel level of a target area to be monitored, or the
like, the motion of a person who is closer to the monitoring camera
is large and hence a fine pattern or the like of the person can be
observed while the motion of another person who is farther from the
monitoring camera is small and hence a fine pattern or the like of
the other person collapses and cannot be observed. Therefore, while
a comparison, as to a region closer to the monitoring camera,
between the current picture and the immediately preceding picture
shows that there are a large number of differences between them, a
comparison, as to a region farther from the monitoring camera,
between the current picture and the immediately preceding picture
shows that there are a smaller number of differences between them.
As a result, there occurs a phenomenon in which while the motion of
a person closer to the monitoring camera is easily detected, the
motion of another person farther from the monitoring camera is hard
to detect. Therefore, there arises a problem that there is a case
in which the motion of a person farther from the monitoring camera
is not detected.
[0031] In order to solve the above-mentioned problem, there can be
considered a method of increasing the sensitivity for motion
detection (making it possible to determine that "there is a motion"
even though the number of differences is small). However, in this
case, there is an increase in the frequency of erroneous detection
to determine that "there is a motion" even from a small difference
which is not a person's motion.
[0032] Further, even when the brightness within a picture changes
due to a flicker of a luminaire, reflection or fluctuations of
light, or the like, and even when a target has no motion in a
partial region within a picture, there is a possibility that the
frequency of erroneous detection to determine that "there is a
motion" increases due to a change of the signal level.
[0033] Further, in order to solve the above-mentioned problem,
there can be considered a method of lowering the sensitivity for
motion detection (making it possible not to determine that "there
is a motion" even though the signal level changes a little).
However, in this case, there is an increase in omissions in
detection including non-detection of the motion of a person and the
motion of a target to be monitored.
[0034] Next, a solution in this Embodiment 1 to the above-mentioned
problem will be described hereafter.
[0035] FIG. 1 is a block diagram showing the configuration of the
whole of a monitoring system including a monitoring camera 1 in
accordance with Embodiment 1 of the present invention. Referring to
FIG. 1, reference numeral 1 denotes the monitoring camera,
reference numeral 2 denotes a network, reference numeral 3 denotes
a PC for monitoring, reference numeral 4 denotes a network
recorder, reference numeral 5 denotes a monitor, and reference
numeral 6 denotes a PC for setting. The network 2 can be the
Internet, a local network, or another network. The present
invention is not limited by the form of the network 2. Further,
although the explanation is made in Embodiment 1 with the
configuration shown in FIG. 1, the monitoring camera 1, the PC for
monitoring 3 and the network recorder 4 can be connected to one
another via not the network 2, but a HUB. As an alternative, the
monitoring camera 1 and the network recorder 4 can be connected
directly to each other, the monitoring camera 1, the PC for
monitoring 3 and the network recorder 4 can be connected directly
to one another, or the monitoring camera 1, the network recorder 4
and the PC for monitoring 3 can be connected to one another by
using another method. The present invention is not limited by the
form of the connection. More specifically, although the following
explanation will be made by assuming that the network recorder 4 is
connected via the network 2, the network recorder 4 can be
alternatively connected not via the network 2.
[0036] The monitoring camera 1, the PC for monitoring 3 and the
network recorder 4 are connected to the network 2. A picture
acquired by performing shooting on a target area by using the
monitoring camera 1 and motion detection information based on that
picture (information indicating the result of motion determination,
which is also referred to as a motion detection result 22 from here
on) are monitored by the PC for monitoring 3, recorded in the
network recorder 4, and displayed by the monitor 5.
[0037] Camera settings of the monitoring camera 1 (a setting of a
setting parameter 21 which will be described below, and another
setting for controlling various operations of the monitoring camera
1) are performed by the PC for setting 6. Referring to FIG. 1, the
PC for setting 6 is connected directly to the monitoring camera 1.
As an alternative, the PC for setting can be connected to the
monitoring camera via a HUB or the network 2. Although the
explanation will be made hereafter by assuming that the camera
settings of the monitoring camera 1 are performed by the PC for
setting 6, the camera settings can be alternatively performed by
either the PC for monitoring 3 or the network recorder 4, and the
present invention is not limited to those examples.
[0038] FIG. 2 is a block diagram showing the configuration of the
monitoring camera 1 in accordance with Embodiment 1 of the present
invention. Referring to FIG. 2, reference numeral 10 denotes a
picture shooter, reference numeral 11 denotes a picture input unit,
reference numeral 12 denotes a picture processing unit, reference
numeral 13 denotes an input output unit (a detection sensitivity
acquirer, an outputter), reference numeral 14 denotes a storage
unit, and reference numeral 15 denotes a motion detector. Further,
reference numeral 21 denotes the setting parameter, reference
numeral 22 denotes the motion detection result, and reference
numeral 31 denotes a processed picture.
[0039] The picture shooter 10 performs shooting on a target area to
be monitored to acquire a picture. The picture acquired by this
picture shooter 10 is inputted to the picture input unit 11. The
picture input unit 11 then outputs this picture to the picture
processing unit 12, and the picture processing unit 12 performs
picture processing on this picture to acquire a processed picture
31. As the picture processing performed by the picture processing
unit 12, a signal level correction, such as a gamma correction, a
color correction, a white balance process, a noise reduction
process, picture enlargement and reduction, a flicker correction, a
sensitization process, an image synthesizing process, edge
enhancement, a rendering process, and so on are provided. The
picture processing unit performs the picture processing by
performing either one of those processes or a combination of two or
more of those processes. As an alternative, the picture processing
unit can perform other well-known picture processing or perform
picture processing by combining the other well-known picture
processing with the above-mentioned processing, and the present
invention is not limited to those examples.
[0040] Further, the input output unit 13 has a function (the
detection sensitivity acquirer) of acquiring the setting parameter
21 from the PC for setting 6, and causes the storage unit 14 to
store the setting parameter 21 acquired thereby. The setting
parameter 21 in accordance with Embodiment 1 is the sensitivity
parameter (detection sensitivity), for each block, used for making
the motion detector 15 perform motion determination. This
sensitivity parameter for each block will be described below.
[0041] When the setting parameter 21 is read from the monitoring
camera 1 to outside the monitoring camera, the input output unit 13
reads the setting parameter 21 stored in the storage unit 14 and
outputs that setting parameter 21 to the PC for setting 6 or the PC
for monitoring 3.
[0042] Further, while the processed picture 31 acquired by the
picture processing unit 12 is outputted, as an inputted picture for
the motion determination, to the motion detector 15, the processed
picture is outputted, as a reference picture for the motion
determination which will be performed the next time or later, to
the storage unit 14. The storage unit 14 stores and manages the
processed picture 31 (the reference picture) inputted thereto in
the inputted order (in time order).
[0043] The motion detector 15 reads the setting parameter 21 (the
sensitivity parameter for each block) stored in the storage unit 14
first. The motion detector 15 also acquires, as the inputted
picture, the processed picture 31 from the picture processing unit
12, and acquires, as the reference image, a processed picture 31
(this picture can be a processed picture 31 of an immediately
preceding frame) which precedes in time that inputted picture from
the storage unit 14. The motion detector 15 has a function (a block
partitioner) of partitioning each of the inputted and reference
pictures which are acquired thereby into a plurality of blocks, and
partitions each of those inputted and reference pictures into a
plurality of blocks. The motion detector 15 then performs the
motion determination on each of the blocks according to the
corresponding sensitivity parameter read thereby by using the
inputted picture and the reference image which are partitioned. A
motion detection result 22 acquired by this motion detector 15 is
outputted to the input output unit 13 and is also outputted to the
PC for monitoring 3 by the input output unit (the outputter)
13.
[0044] In Embodiment 1, the pictures outputted from the picture
processing unit 12 and the storage unit 14 to the motion detector
15 are the current inputted picture and the reference picture used
for performing the motion determination. The reference picture can
be a processed picture 31 inputted immediately before the inputted
picture, or can be another processed picture 31 already inputted at
a time different from the time when the processed picture 31 is
inputted immediately before the inputted picture. More
specifically, the reference picture has only to be a processed
picture 31 which can be used for performing the motion
determination. The present invention is not limited to those
examples.
[0045] Further, in FIG. 2, the case in which the inputted picture
is outputted from the picture processing unit 12 directly to the
motion detector 15 is shown. In contrast with this, instead of
outputting the inputted picture from the picture processing unit 12
directly to the motion detector 15, the inputted picture and the
reference picture can be outputted from the storage unit 14 to the
motion detector 15. More specifically, the inputted picture and the
reference image have only to be inputted to the motion detector 15,
and the present invention is not limited to those examples.
[0046] Further, although it is assumed in Embodiment 1 that the
partitioning of each of the inputted and reference pictures into
blocks is performed by the motion detector 15, the function (the
block partitioner) can be alternatively disposed in the storage
unit 14. As an alternative, the function (the block partitioner) of
partitioning the inputted picture into blocks can be disposed in
the picture processing unit 12. More specifically, what is
necessary is just to be able to carry out the block partitioning in
such a way that the motion determination can be performed on each
block, and the present invention is not limited by the functional
unit to partition each of the pictures into blocks and the block
partitioning method. Further, the state of the block partitioning
(e.g., the number of blocks after the partitioning, and the range
of each block) can be notified to the PC for setting 6 or the PC
for monitoring 3 via the input output unit 13.
[0047] FIG. 3 is a diagram showing the picture partitioning for the
motion determination. FIG. 3 shows a case in which a picture is
partitioned into 48 blocks. The motion detector 15 performs the
motion determination on each of these blocks after the
partitioning. Although a picture is partitioned into 48 blocks in
the example shown in FIG. 3, a picture can be alternatively
partitioned into 12 blocks, 16 blocks, 100 blocks, or an arbitrary
number of blocks, and the present invention is not limited to those
examples.
[0048] The target which is partitioned into blocks having a uniform
size is assumed to be one frame which constructs one screen.
Further, in the case in which one frame consists of two fields,
like in the case of an interlaced picture, one frame can be
partitioned into blocks having a uniform size or each of the fields
can be partitioned into blocks having a uniform size, and the
present invention is not limited to those examples. Although
multiple frames can be handled as a group and partitioned into
blocks having a uniform size, the memory capacity to store blocks
required to perform the motion determination can be reduced through
the partitioning of each frame or each field into blocks having a
uniform size, as compared with a case of storing the blocks of
multiple frames.
[0049] Further, the sensitivity parameter used for the motion
determination by the motion detector 15 is a parameter used for
determining the presence or absence of a motion, with respect to a
motion determination index value calculated from the inputted
picture and the reference picture. As the motion determination
index value, for example, the number of pixels each of which is
determined, from a comparison between the inputted picture and the
reference picture, to have a difference between them is used.
[0050] At that time, a change of each pixel between the inputted
picture and the reference picture is calculated by, instead of
tracing to which coordinates the pixel in question has moved,
calculating a change of the pixel value at the same coordinates in
the same block. As a result, the presence or absence of a motion
can be determined for the block in question. More specifically,
instead of determining how each pixel in an object which is a
target has moved, whether or not a moving target is staying in the
block in question is determined.
[0051] In the trace determination of an object, a memory that
stores frames within a time period which is the target for tracing
is needed. In contrast with this, in the case of using the method
of determining whether or not a moving target is staying in the
block in question, only a memory that stores a reference picture
corresponding to the inputted picture is needed and the memory
capacity can be reduced as compared with the trace determination of
an object. Further, in the case of detection of a suspicious
person, while it is important to detect how a suspicious person has
moved, it is important to perform determination of whether or not a
suspicious person is staying (there is a moving target in the block
in question) and determination of the presence or absence of a
suspicious person can be performed with a relatively smaller memory
capacity.
[0052] FIG. 4 is a diagram showing a relation between the motion
determination index value and the sensitivity parameter, the
relation being set according to variations in the distance from the
monitoring camera 1 to a target area (the distance to the target
area with the installation position of the monitoring camera 1
being defined as a reference point), and FIG. 4(a) is a diagram
showing a case of using a single sensitivity parameter and FIG. 4
(b) is a diagram showing a case of using a plurality of sensitivity
parameters.
[0053] As shown in FIG. 4, there is a tendency that, even when
persons are moving similarly, the motion determination index value
becomes large for a person who is moving in a region closer to the
monitoring camera 1, while the motion determination index value
becomes small for another person who is moving in a region farther
from the monitoring camera 1.
[0054] Hereafter, a case will be considered in which the motion
detector 15 determines that "there is a motion" when the motion
determination index value is equal to or greater than a sensitivity
parameter, whereas the motion detector determines that "there is no
motion" when the motion determination index value is less than the
sensitivity parameter.
[0055] At that time, if performing the motion determination on a
picture by using a single sensitivity parameter, as shown in FIG.
4(a), the motion detector determines that, as to the motion
determination index value acquired for an object closer to the
monitoring camera 1, "there is a motion", while the motion detector
determines that, as to the motion determination index value
acquired for an object farther from the monitoring camera 1, "there
is no motion."
[0056] To solve this problem, in accordance with the present
invention, as shown in FIG. 4 (b), for each picture, sensitivity
parameters (first and second sensitivity parameters) which are
different between a region closer to the monitoring camera 1 and a
region farther from the monitoring camera 1 are set. As a result,
when persons are moving similarly, not only as to the motion
determination index value acquired for a person moving in a region
closer to the monitoring camera 1, but also as to the motion
determination index value acquired for another person moving in a
region farther from the monitoring camera 1, the motion detector
can determine that "there is a motion" according to the value.
[0057] As mentioned above, in accordance with the present
invention, in order to perform the motion determination in
consideration of the difference in the detection sensitivity, the
difference being dependent upon the distance from the monitoring
camera 1 to the target area, the motion detector 15 (the block
partitioner) partitions each picture into a plurality of blocks and
the PC for setting 6 sets a sensitivity parameter for each of the
blocks. At that time, the PC for setting 6 sets the sensitivity
parameter according to the distance between the target area in each
of the blocks and the monitoring camera 1. More specifically,
because it is difficult to detect the motion of a target staying
farther from the monitoring camera 1 as compared with that of a
target staying closer to the monitoring camera 1, the PC for
setting sets the sensitivity parameter for a region farther from
the monitoring camera 1 to a value which makes it easier to detect
a motion, as compared with the sensitivity parameter for a region
closer to the monitoring camera.
[0058] FIG. 5 is a diagram showing an example of a monitor picture
acquired by the monitoring camera 1. In the example shown in FIG.
5, in the picture, a person is seen in a region (a zone B) closer
to the monitoring camera 1 and another person is seen in a region
(a zone A) farther from the monitoring camera 1. In this case, the
PC for setting sets the sensitivity parameter for each block
included in the zone A to a value which makes it easier to detect a
motion (in such a way as to increase the sensitivity), as compared
with the sensitivity parameter for each block included in the zone
B. As a result, a situation in which it is difficult to detect the
motion of a person staying farther from the monitoring camera 1 can
be eliminated.
[0059] The zones A and B in the example shown in FIG. 5 can be set
up by a user by making a judgment on the basis of the angle of view
of the monitoring camera 1 to perform a zone division, or can be
determined by the monitoring camera 1 by using an automatic
distance measurement method to measure distances and perform a zone
division, and the present invention is not limited to those
examples.
[0060] Further, the PC for setting 6 can set the sensitivity
parameter according to an influence of light (the presence or
absence of a flicker, fluctuations or the like of light) upon each
block. At that time, the PC for setting 6 sets the sensitivity
parameter for a block in which, for example, a fluorescent lamp is
seen in the picture to a value which makes it difficult to detect a
motion (in such a way as to lower the sensitivity) as compared with
the sensitivity parameter for a block in which no fluorescent lamp
is seen in the picture. Further, the PC for setting 6 sets the
sensitivity parameter for a block in which, for example, reflected
light from a luminaire is seen in the picture to a value which
makes it difficult to detect a motion (in such a way as to lower
the sensitivity) as compared with the sensitivity parameter for a
block in which no reflected light from a luminaire is seen in the
picture. Further, also when no fluorescent lamp is seen in the
picture, but a light source having an influence (erroneous
detection) upon motion detection due to a flicker or fluctuations
of light is seen in the picture, the PC for setting sets the
sensitivity parameter for a block in which that light source is
seen in the picture to a value which makes it difficult to detect a
motion as compared with the sensitivity parameter for a block in
which that light source is not seen in the picture, in the same way
as above.
[0061] As mentioned above, even in a case in which the picture
contains a region in which erroneous detection of a motion may be
performed under the influence of light, by setting the sensitivity
parameter for each block in that region to a value which makes it
difficult to detect a motion, erroneous detection can be
suppressed. In this case, although the detection accuracy decreases
only in such a block as above, when a target to be monitored (e.g.,
a person) has a size extending over a plurality of blocks,
detection of the motion of the target can be performed in blocks
other than that block.
[0062] In Embodiment 1, the case in which the setting parameter 21
set by the PC for setting 6 is a sensitivity parameter used for
performing the motion determination is shown. In contrast with
this, the PC for setting 6 can alternatively set an immediate
sensitivity parameter for each block which is a unit for the motion
determination within the picture.
[0063] As an alternative, the PC for setting 6 can set both a base
sensitivity parameter which is a base for the entire picture, and a
sensitivity offset for each block which is a unit for the motion
determination, and the monitoring camera 1 (the motion detector 15)
can calculate the sensitivity parameter for each block on the basis
of those base sensitivity parameter and sensitivity offset.
Further, the PC for setting can set the sensitivity parameter by
performing a grouping setting (a zone setting) or the like. As an
alternative, the PC for setting 6 or the network recorder 4 can
perform camera settings and the monitoring camera 1 can perform the
motion determination.
[0064] More specifically, the value of the sensitivity parameter
can be arbitrary as long as the sensitivity parameter has a value
which makes it possible to perform the motion determination, and
the method of setting the sensitivity parameter does not limit the
present invention. The PC for setting 6 or the network recorder 4
can alternatively perform camera settings and the network recorder
4 can alternatively receive the picture acquired by the monitoring
camera 1 and perform the motion determination on the basis of the
picture inputted thereto, and the device that performs camera
settings and the device that performs the motion determination do
not limit the present invention. An example of the configuration of
the network recorder 4 in the case of inputting the picture
acquired by the monitoring camera 1 to the network recorder 4, and
performing the motion determination by means of the network
recorder 4 on the basis of the inputted picture will be explained
by using FIG. 2. In this case, the network recorder 4 has only to
have, for example, the input output unit 13 having the function
(the detection sensitivity acquirer) of acquiring the setting
parameter, the storage unit 14 that stores the setting parameter,
the motion detector 15 and the input output unit 13 that outputs a
motion detection result, which are shown in FIG. 2. As an
alternative, the picture acquired by the monitoring camera 1 can be
inputted to the PC for monitoring 3, and the PC for monitoring 3
can perform the motion determination on the basis of the picture
inputted thereto. In this case, the PC for monitoring 3 has only to
have, for example, the input output unit 13 having the function
(the detection sensitivity acquirer) of acquiring the setting
parameter, the storage unit 14 that stores the setting parameter,
the motion detector 15 and the input output unit 13 that outputs a
motion detection result, which are shown in FIG. 2.
[0065] In the case in which the monitoring camera 1 (the motion
detector 15) calculates the sensitivity parameter for each block on
the basis of the base sensitivity parameter and the sensitivity
offset, the monitoring camera can calculate the sensitivity
parameter for each block by adding or subtracting the sensitivity
offset to or from the base sensitivity parameter. As an
alternative, the monitoring camera can calculate the sensitivity
parameter for each block by multiplying or dividing the base
sensitivity parameter by the sensitivity offset (a coefficient). As
an alternative, the monitoring camera can calculate the sensitivity
parameter for each block by performing an arbitrary arithmetic
operation, other than the above-mentioned arithmetic operations, on
the base sensitivity parameter and the sensitivity offset, and the
present invention is not limited to those examples.
[0066] For example, when the arithmetic operation is expressed by a
function f (x, y) (in this function, x: the base sensitivity
parameter, y: the sensitivity offset), and the function is one of
the four operations of arithmetic, the function is given by either
one of the following equations: (1) the calculated sensitivity=the
base sensitivity parameter+the sensitivity offset; (2) the
calculated sensitivity=the base sensitivity parameter-the
sensitivity offset; (3) the calculated sensitivity=the base
sensitivity parameter.times.the sensitivity offset; and (4) the
calculated sensitivity=the base sensitivity parameter .epsilon. the
sensitivity offset.
[0067] As an alternative, when the function is an exponential
function (f(x, y)=x y) (the (sensitivity offset)-th power of the
base sensitivity parameter), the function is given by the
calculated sensitivity=the base sensitivity parameter the
sensitivity offset.
[0068] As an alternative, the function can be f(x, y)=x+3y+2, f(x,
y)=x.times.2y-1, or another one. The function can be alternatively
a sensitivity parameter which is updated by performing an
arithmetic operation (a function arithmetic operation) based on the
base sensitivity parameter and the sensitivity offset, or an
arithmetic operation (i.e., f(x, y)=x) which yields the calculated
sensitivity equal to the base sensitivity parameter.
[0069] In Embodiment 1, the sensitivity parameter is set as a
parameter used for performing the motion determination on the
motion determination index value which is calculated on the basis
of the inputted picture and the reference picture. However, the
sensitivity parameter can have an arbitrary range of values or an
arbitrary value as long as the sensitivity parameter is a one used
for performing the motion determination, and the present invention
is not limited to those examples. As a concrete example of the case
in which the sensitivity parameter has a range of values, there is
a case in which it is determined that "there is a motion" when the
motion determination index value is equal to or greater than a
first value or equal to or less than a second value, or the like.
In this case, the first and second values can be set as the
sensitivity parameter, or a range (a beltlike range) less than the
first value and greater than the second value can be set as the
sensitivity parameter.
[0070] Further, the sensitivity parameter can have a value which is
used for the processing on one screen or one frame, or the same
sensitivity parameter can be used for the processing on subsequent
screens (a plurality of screens in the time axis direction) or
subsequent frames (a plurality of frames in the time axis
direction) until a change is made to the sensitivity parameter.
[0071] Further, in Embodiment 1, the case in which the motion
determination index value is defined to be the number of pixels
each of which is determined, from a comparison between the inputted
picture and the reference picture, to have a difference between
them is shown. In contrast with this, the sum total (the difference
absolute value sum) of the differences in the values of pixels
between the inputted picture and the reference picture (the
differences in the pixel values at the same pixel positions) can be
alternatively defined as the motion determination index value. As
an alternative, each of the inputted and reference pictures can be
partitioned into blocks, and the differences (the difference
absolute value sum) in the values of pixels between the inputted
picture and the reference picture in each block can be defined as
the motion determination index value. Further, the differences in
the pixel values can be the differences in luminance signals or the
differences in chrominance signals. As an alternative, the
difference in edge component which is acquired through calculation
of edge components can be defined as the motion determination index
value. The calculation of edge components can be performed by using
any of well-known methods or any combination of two or more of
those methods. More specifically, the motion determination index
value can be arbitrary as long as the motion determination index
value shows the difference in picture feature between the inputted
picture and the reference picture, and the present invention is not
limited to those examples. Thus, when the motion determination
index value is determined to be an index value showing the
difference in picture feature between the inputted picture and the
reference picture, what is necessary is just to set the sensitivity
parameter for a region farther from the monitoring camera 1 to a
value lower than the sensitivity parameter for a region closer to
the monitoring camera. As a result, a motion in a region farther
from the monitoring camera can be detected more easily as compared
with that in a region closer to the monitoring camera.
[0072] When partitioning each picture into blocks, only a region
which is a target for the motion determination within the picture
can be partitioned into blocks and each of the blocks can be set as
a target for the motion determination. Further, a specified block
in the blocks into which the picture is partitioned can be set to
be a determination exclusion block which is not a target for the
motion determination. More specifically, what is necessary is just
to be able to carry out the motion determination on each block
which is a target for the motion determination, and the present
invention is not limited to those examples. For example, a central
portion of each picture and a peripheral portion surrounding the
central portion can be set as a target for the motion
determination, and right and left edge portions and upper and lower
edge portions of the picture can be set as an outside of the target
for the motion determination. Further, some of the blocks for the
motion determination into which the picture is partitioned can be
set as an outside of the target for the motion determination (a
so-called masked region).
[0073] Next, the operation of the monitoring system configured as
above will be explained. FIG. 6 is a flow chart showing the
operation of the whole of the monitoring system in accordance with
Embodiment 1 of the present invention. The operation of the
monitoring camera 1 will be explained hereafter. The monitoring
camera 1 can shoot a video and output a picture and the network
recorder 4 and the PC for monitoring 3 can perform subsequent
processes, and the PC for setting 6 and the network recorder 4 can
perform camera settings. The present invention is not limited to
that example.
[0074] In the operation of the whole of the monitoring system,
first, the picture input unit 11, in step ST601, receives the
picture (a picture signal or picture data) which the picture
shooter 10 has acquired by performing shooting on a target area,
and outputs the picture to the picture processing unit 12.
[0075] The picture processing unit 12 then, in step ST602, acquires
a processed picture 31 by performing the picture processing on the
inputted picture. The picture processing unit 12 then outputs the
processed picture 31 acquired thereby to the storage unit 14 and
the motion detector 15.
[0076] The motion detector 15 then, in step ST603, acquires, as an
inputted picture, the processed picture 31 from the picture
processing unit 12.
[0077] The input output unit 13 then, in step ST604, acquires the
setting parameter 21 from the PC for setting 6, and causes the
storage unit 14 to store the setting parameter. Instead of storing
the setting parameter 21 acquired by the input output unit 13 in
the storage unit 14, the input output unit can output the setting
parameter directly to the motion detector 15.
[0078] The motion detector 15 then, in step ST605, reads the
setting parameter 21 stored in the storage unit 14.
[0079] In a case in which the base sensitivity parameter is stored
in advance in the storage unit 14, and the setting parameter 21
acquired by the input output unit 13 is the sensitivity offset set
for each block, the motion detector 15 reads those base sensitivity
parameter and sensitivity offset. When no input of the sensitivity
offset set for each block is performed, the motion detector can
alternatively read only the base sensitivity parameter. The motion
detector 15 can alternatively hold the base sensitivity parameter
in advance.
[0080] The motion detector 15 then, in step ST606, acquires, as the
reference picture, a processed picture 31 which has been inputted
to and stored in the storage unit 14 (a preceding frame) before in
time the processed picture 31 (the inputted picture) acquired in
step ST603.
[0081] The motion detector 15 then, in step ST607, performs the
motion determination process by using the inputted picture acquired
in step ST603, the reference picture acquired in step ST606, and
the setting parameter 21 read in step ST605. The motion detection
result 22 acquired by this motion detector 15 is outputted to the
input output unit 13. A detailed explanation of the motion
determination process performed by the motion detector 15 will be
made below by using FIG. 7.
[0082] The input output unit 13 then, in step ST608, outputs the
motion detection result 22 acquired by the motion detector 15 to
the PC for monitoring 3. The input output unit 13 can be
alternatively configured in such a way as to, only when the motion
detection result 22 shows that "there is a motion", output the
motion detection result 22 to the PC for monitoring 3.
[0083] Next, the motion determination process performed by the
motion detector 15 will be explained by using FIG. 7. Hereafter,
the case in which the motion detector acquires, as the setting
parameter 21, the sensitivity parameter set for each block from the
storage unit 14 will be shown.
[0084] In the motion determination process by the motion detector
15, the motion detector first, in step ST701, partitions each of
the inputted and reference pictures into a plurality of blocks. As
an alternative, the partitioning of each of the inputted and
reference pictures into a plurality of blocks can be performed by
either the picture processing unit 12 or the storage unit 14. In
that case, the motion detector 15 acquires the inputted picture on
which the block partitioning is performed and the reference picture
on which the block partitioning is performed. The motion detector
15 can be further configured in such a way as to read data about
each block from either the picture processing unit 12 or the
storage unit 14 (addressing).
[0085] The motion detector then, in step ST702, calculates the
motion determination index value for each block on the basis of the
inputted picture and the reference picture.
[0086] Next, the motion detector, in step ST703, performs the
motion determination. As a concrete example, the motion detector
determines whether or not the motion determination index value is
equal to or greater than the corresponding sensitivity parameter
for each block. At that time, when the motion determination index
value is equal to or greater than the corresponding sensitivity
parameter, the motion detector shifts to step ST704 and determines
that "there is a motion." In contrast, when the motion
determination index value is less than the corresponding
sensitivity parameter, the motion detector shifts to step ST705 and
determines that "there is no motion."
[0087] The motion detector then, in step ST706, outputs the motion
detection result 22 to the input output unit 13 on a per block
basis. The motion detector can be alternatively configured in such
a way as to, only when the motion detection result 22 shows "there
is a motion", output that motion detection result 22.
[0088] As mentioned above, because the monitoring system in
accordance with this Embodiment 1 is configured in such a way as to
change and set the sensitivity parameter for each block according
to a motion detection condition dependent upon the distance from
the monitoring camera 1 of the monitoring system to the target
area, a motion detection condition dependent upon the influence of
light within the picture, etc., and to perform the motion
determination on each block, the monitoring system can prevent
erroneous motion detection and omissions in motion detection in
each of blocks into which the picture is partitioned, with a simple
and low-cost configuration. As a result, the monitoring camera 1
can be prevented from providing an erroneous warning and omitting
any warning.
[0089] Further, because the monitoring camera 1 of the monitoring
system performs the motion determination by using the sensitivity
parameter different for each block, a plurality of monitoring
stations (PCs for monitoring 3) do not have to be disposed, unlike
in the case of using a conventional technology, and therefore the
cost required to configure the monitoring system can be reduced.
Further, the number of blocks into which the picture is partitioned
is not limited to the order of monitoring stations. Further,
because the monitoring system performs the motion determination by
using the sensitivity parameter different for each block, any
transfer of large-volume data from the monitoring camera 1 to any
monitoring station becomes unnecessary.
[0090] Further, in the example shown in FIG. 3, the picture is
partitioned into blocks having a uniform size. In this case, blocks
having a uniform size satisfy at least one of, for example, the
following conditions: (1) the number of pixels contained in each
block is the same; (2) the number of pixels in a vertical direction
contained in each block is the same and the number of pixels in a
horizontal direction contained in each block is the same, where the
number of pixels in a vertical direction may differ from the number
of pixels in a horizontal direction; and (3) the shape and the size
of each block is congruent (the shape includes a shape which is not
a rectangle). Blocks having a uniform size can be alternatively
defined according to another condition other than the
above-mentioned conditions, and the present invention is not
limited to those examples.
[0091] Through the partitioning of the picture into blocks having a
uniform size, the control of the motion determination on each block
is facilitated. This is because the motion determination index
value calculated for each block can be compared with the motion
determination index value calculated for another block through the
partitioning of the picture into blocks having a uniform size (the
motion determination index value has only to be a value indicating
a pixel feature of the corresponding block by using pixel value
information about the inside of the block, such as the sum total,
the multiplication or the accumulation of the pixel values in the
block, and each pixel value has only to be a luminance signal
value, a chrominance signal value or an RGB signal value of the
corresponding pixel, or another value indicating the corresponding
pixel).
[0092] More specifically, as to blocks in a region close to the
monitoring camera 1, whether or not there is a motion is determined
by using the same sensitivity parameter. On the other hand, there
is a case in which if the distance from the monitoring camera 1
differs, even though objects have the same motion, the motion
determination index values acquired for corresponding blocks are
different from each other. Although the motion determination index
value is made to differ according to the difference in the distance
from the monitoring camera 1 through the partitioning of the
picture into blocks having a uniform size, the partitioning into
blocks having a uniform size makes it easy to set the sensitivity
parameter according to the difference in the distance from the
monitoring camera 1.
[0093] Therefore, for blocks in which the motion determination
index values differ according to the difference in the distance
from the monitoring camera 1, among the blocks having a uniform
size into which the picture is partitioned, the motion
determination is performed by using different sensitivity
parameters, respectively. As a result, even when the distance from
the monitoring camera 1 differs, the motion determination can be
performed on similar motions of objects.
[0094] For example, when the speed at which a person is walking in
a region closer to the monitoring camera 1 is the same as the speed
at which another person is walking in a region farther from the
monitoring camera 1 within the screen, the motion of the person
walking in the region closer to the monitoring camera 1 is
observed, within the screen, to be a motion which is relatively
larger than the motion of the other person walking in the region
farther from the monitoring camera 1. Therefore, the sensitivity
parameter is specified independently according to the difference in
the distance from the monitoring camera 1. As a result, it can be
determined for both blocks showing the region closer to the
monitoring camera 1 and blocks showing the region farther from the
monitoring camera 1 that "there is a motion."
[0095] If the same sensitivity parameter is specified for any of
the blocks having a uniform size irrespective of the distance from
the monitoring camera 1, there may be a case in which, for example,
it is determined for blocks showing the region closer to the
monitoring camera 1 that "there is a motion," while the motion
index value becomes small relatively for blocks showing the region
farther from the monitoring camera 1, and it is determined for the
blocks that "there is no motion." In contrast with this, by
specifying the sensitivity parameter independently according to the
difference in the distance from the monitoring camera 1, as
mentioned above, the presence or absence of a motion of an object
within each block can be determined.
Embodiment 2
[0096] First, a problem to be solved by Embodiment 2 will be
explained. The basic configuration of a monitoring camera 1 in
accordance with Embodiment 2 is the same as that of the monitoring
camera 1 in accordance with Embodiment 1 shown in FIG. 2, and the
explanation will be made by focusing on a different portion.
Although the following explanation will be made by focusing on a
case in which the monitoring camera 1 performs motion
determination, also in the case of Embodiment 2, a picture acquired
by the monitoring camera 1 can be inputted to a network recorder 4
and the network recorder 4 can alternatively perform motion
determination on the basis of the picture inputted thereto, like in
the case of Embodiment 1. Further, also in the case of Embodiment
2, the picture acquired by the monitoring camera 1 can be
alternatively inputted to a PC for monitoring 3 and the PC for
monitoring 3 can alternatively perform motion determination on the
basis of the picture inputted thereto, like in the case of
Embodiment 1.
[0097] In Embodiment 1, a motion detector 15 performs the motion
determination on the basis of the inputted picture and a reference
picture which are acquired from a picture processing unit 12 and a
storage unit 14. When a person is moving as shown in FIG. 8, there
exist, within a certain frame time period, a time period during
which the person's motion is continuing without stopping
momentarily, and a time period during which the person stops its
motion and is at rest. On the other hand, as shown by "*" in the
picture of FIG. 8, there may be a flicker or fluctuations of light
on the floor due to reflection from a luminaire, or the like, or
blinking of a light source such as an LED mounted in equipment. In
the case of using a method of determining whether or not there is a
motion from a signal change in such a situation, it may be
determined erroneously that "a motion is detected" on the basis of
a signal change, when only a single set of an inputted picture and
a reference picture is used to perform the motion
determination.
[0098] To solve this problem, in accordance with Embodiment 2, the
motion detector 15 determines whether or not there is a block in
which a motion has continued for a set time period. Concretely, the
motion detector 15 stores, as a primary detection result, a
detection result acquired through the motion determination using a
sensitivity parameter shown in Embodiment 1, and performs secondary
motion determination of determining whether or not there is a block
which is determined in the primary detection result to be a one, in
consecutive frames, in which "there is a motion." This process is
aimed at suppressing erroneous detection in the motion detection in
an environment in which erroneous detection may occur in the motion
detection due to a flicker, fluctuations, or the like of light from
a luminaire or the like.
[0099] When there is a block which is determined in the primary
detection result to be a one, in consecutive frames, in which
"there is a motion", that motion detection shows that the motion is
a one of a moving object such as a person, and makes it possible to
distinguish the motion from a flicker and fluctuations of light and
blinking of a light source.
[0100] FIG. 9 is a diagram showing an example of determination of
the continuity of a motion which is performed by the monitoring
camera 1 in accordance with Embodiment 2 of the present invention,
and FIG. 9(a) shows a temporal change of the motion determination
index value which is caused by an influence of light and FIG. 9(b)
is a diagram showing a temporal change of the motion determination
index value which is caused by a motion of a moving object.
[0101] A moving object, such as a person, an animal, a car or a
train, has continuity in its motion. Therefore, as shown in FIG.
9(b), when a moving object is moving, its motion determination
index value keeps a large value in the meantime, but when the
moving object stops its motion and is at rest, the motion
determination index value becomes small in the meantime. In
contrast, in the case of a flicker or fluctuations of light from a
luminaire or the like, its motion determination index value becomes
large or small during a short time period, as shown in FIG.
9(a).
[0102] Therefore, for example, a consecutive frame number is used
as a criterion (a set time period) by which to determine whether or
not there is continuity in a motion, as shown in FIG. 9. When the
consecutive frame number is set to 3, because the time period
during which the motion determination index value in each block
which is a target for the determination is equal to or greater than
a sensitivity parameter corresponds to every other frame, not every
three or more consecutive frames, in the example of FIG. 9(a), it
is determined that there is no continuity in the motion (there is
no motion). In contrast, because the time period during which the
motion determination index value in each block which is a target
for the determination is equal to or greater than the sensitivity
parameter corresponds to eight consecutive frames, i.e., three or
more consecutive frames, in the example of FIG. 9(b), it is
determined that there is continuity in the motion (there is a
motion). By thus determining whether or not there is continuity in
time in the motion determination index value according to the
consecutive frame number, whether or not a motion is a one of a
moving object can be determined.
[0103] Although the case in which the consecutive frame number is
set to 3 is shown in the example of FIG. 9, the consecutive frame
number can be arbitrary as long as the number has a value which
makes it possible to perform the motion determination on a moving
object, and the present invention is not limited to that example.
Further, although the case in which the consecutive frame number is
used is as the criterion (the set time period) by which to
determine whether or not there is continuity in time in the motion
determination index value is shown in the example of FIG. 9, a unit
time can be alternatively used and the present invention is not
limited to those examples.
[0104] Further, information showing the consecutive frame number
can be acquired, as a setting parameter 21, from a PC for setting 6
by an input output unit 13 (a set time period acquirer). As an
alternative, the PC for setting 6 can set the consecutive frame
number according to the presence or absence of an influence of
light.
[0105] Next, the motion determination process performed by the
motion detector 15 in accordance with Embodiment 2 will be
explained by using FIG. 10. Hereafter, a case in which the motion
detector acquires a setting parameter 21 (a sensitivity parameter)
from the storage unit 14 will be shown.
[0106] In the motion determination process by the motion detector
15, the motion detector first, in step ST1001, partitions each of
inputted and reference pictures into a plurality of blocks. As an
alternative, the partitioning of each of the inputted and reference
pictures into a plurality of blocks can be performed by either the
picture processing unit 12 or the storage unit 14. In that case,
the motion detector 15 acquires the inputted picture on which the
block partitioning is performed and the reference picture on which
the block partitioning is performed. The motion detector 15 can be
further configured in such a way as to read data about each block
from either the picture processing unit 12 or the storage unit 14
(addressing).
[0107] The motion detector then, in step ST1002, calculates the
motion determination index value for each block on the basis of the
inputted picture and the reference picture.
[0108] Next, the motion detector, in step ST1003, performs primary
motion determination. As a concrete example, the motion detector
determines whether or not the motion determination index value is
equal to or greater than the corresponding sensitivity parameter
for each block, like that in accordance with Embodiment 1. When the
motion determination index value is equal to or greater than the
corresponding sensitivity parameter, the motion detector shifts to
step ST1004 and determines that "there is a motion (determination
1)." In contrast, when the motion determination index value is less
than the corresponding sensitivity parameter, the motion detector
shifts to step ST1005 and determines that "there is no motion
(determination 1)."
[0109] Next, the motion detector, in step ST1006, performs
secondary motion determination. As a concrete example, the motion
detector determines whether or not the number of consecutive frames
each of which is determined, through the primary motion
determination, to be a one in which "there is a motion
(determination 1)" is equal to or greater than the set consecutive
frame number, on a per block basis. When the number of consecutive
frames each of which is determined as above is equal to or greater
than the set consecutive frame number, the motion detector shifts
to step ST1007 and determines that "there is a motion
(determination 2)." In contrast, when the number of consecutive
frames each of which is determined as above is less than the set
consecutive frame number, the motion detector shifts to step ST1008
and determines that "there is no motion (determination 2)."
[0110] The motion detector then, in step ST1009, outputs the motion
detection result 22 acquired through the secondary motion
determination to the input output unit 13 on a per block basis. The
motion detector can be alternatively configured in such a way as
to, only when the motion detection result 22 shows "there is a
motion (determination 2)", output that motion detection result
22.
[0111] As mentioned above, because the monitoring system in
accordance with this Embodiment 2 is configured in such a way as to
determine whether or not a block having a motion continuing over a
certain number (a set time period) or more of consecutive frames is
included in the blocks into which the picture is partitioned, there
is provided an advantage of being able to distinguish a motion of a
moving object from a flicker and fluctuations of light, blinking of
a light source, etc., in addition to the advantages provided by
Embodiment 1.
[0112] In Embodiment 2, the case in which the secondary motion
determination of determining whether or not there is continuity in
a motion is performed after the primary motion determination is
performed by using the sensitivity parameter for each block is
shown. In contrast with this, the monitoring system can
alternatively perform only the secondary motion determination
without performing the primary motion determination. Also in the
case in which the monitoring system is configured this way, the
monitoring system can prevent erroneous motion detection and
omissions in motion detection in each of the blocks into which the
picture is partitioned, with a simple and low-cost
configuration.
Embodiment 3
[0113] The case in which the consecutive frame number (the set time
period) is used as the criterion for the secondary motion
determination is shown in Embodiment 2. In contrast with this, in
Embodiment 3, a case in which a set block number is used as the
criterion for the secondary motion determination will be shown.
More specifically, a motion detector 15 performs the secondary
motion determination of determining whether or not the number of
blocks each of which is determined in the primary detection result
to be a one in which "there is a motion" is equal to or greater
than the set block number. This process is aimed at identifying the
size of each moving object within the picture.
[0114] When the number of blocks each of which is determined in the
primary detection result to be a one in which "there is a motion"
is equal to or greater than the set block number within the
picture, the motion detector determines that "there is a motion."
For example, when the size of a person staying within the picture
is 4 blocks or more irrespective of its position, and the size of a
small animal is 1 block, as shown in FIG. 11, small moving objects,
such as the small animal, and persons can be distinguished by
setting the set block number to 3.
[0115] The basic configuration of a monitoring camera 1 in
accordance with Embodiment 3 is the same as that of the monitoring
camera 1 in accordance with Embodiment 1 shown in FIG. 2, and the
explanation will be made by focusing on a different portion.
Although the following explanation will be made by focusing on a
case in which the monitoring camera 1 performs the motion
determination, also in the case of Embodiment 3, a picture acquired
by the monitoring camera 1 can be inputted to a network recorder 4
and the network recorder 4 can alternatively perform the motion
determination on the basis of the picture inputted thereto, like in
the case of Embodiment 1. Further, also in the case of Embodiment
3, the picture acquired by the monitoring camera 1 can be
alternatively inputted to a PC for monitoring 3 and the PC for
monitoring 3 can alternatively perform the motion determination on
the basis of the picture inputted thereto, like in the case of
Embodiment 1.
[0116] Further, information indicating the set block number can be
acquired, as a setting parameter 21, from a PC for setting 6 by an
input output unit 13 (a set block number acquirer). The PC for
setting 6 can set the set block number according to the sizes of
moving objects to be distinguished. Although the case in which the
set block number is set to 3 is shown in the example of FIG. 11,
the set block number based on the number of blocks into which the
picture is partitioned or the sizes of moving objects to be
distinguished can be arbitrary, and the present invention is not
limited to those examples.
[0117] Next, the motion determination process performed by the
motion detector 15 in accordance with Embodiment 3 will be
explained by using FIG. 12. Hereafter, a case in which the motion
detector acquires a setting parameter 21 (a sensitivity parameter)
from a storage unit 14 will be shown.
[0118] In the motion determination process by the motion detector
15, the motion detector first, in step ST1201, partitions each of
inputted and reference pictures into a plurality of blocks. As an
alternative, the partitioning of each of the inputted and reference
pictures into a plurality of blocks can be performed by either a
picture processing unit 12 or the storage unit 14. In that case,
the motion detector 15 acquires the inputted picture on which the
block partitioning is performed and the reference picture on which
the block partitioning is performed. The motion detector 15 can be
further configured in such a way as to read data about each block
from either the picture processing unit 12 or the storage unit 14
(addressing).
[0119] The motion detector then, in step ST1202, calculates a
motion determination index value for each block on the basis of the
inputted picture and the reference picture.
[0120] Next, the motion detector, in step ST1203, performs primary
motion determination. As a concrete example, the motion detector
determines whether or not the motion determination index value is
equal to or greater than the corresponding sensitivity parameter
for each block, like that in accordance with Embodiment 1. When the
motion determination index value is equal to or greater than the
corresponding sensitivity parameter, the motion detector shifts to
step ST1204 and determines that "there is a motion (determination
1)" In contrast, when the motion determination index value is less
than the corresponding sensitivity parameter, the motion detector
shifts to step ST1205 and determines that "there is no motion
(determination 1)."
[0121] Next, the motion detector, in step ST1206, performs
secondary motion determination. As a concrete example, the motion
detector determines whether or not the number of blocks each of
which is determined, through the primary motion determination, to
be a one in which "there is a motion (determination 1)" is equal to
or greater than the set block number. When the number of blocks
each of which is determined as above is equal to or greater than
the set block number, the motion detector shifts to step ST1207 and
determines that "there is a motion (determination 2)." In contrast,
when the number of blocks each of which is determined as above is
less than the set block number, the motion detector shifts to step
ST1208 and determines that "there is no motion (determination 2)."
The motion detector then, in step ST1209, outputs the motion
detection result 22 acquired through the secondary motion
determination to the input output unit 13. The motion detector can
be alternatively configured in such a way as to, only when the
motion detection result 22 shows "there is a motion (determination
2)", output that motion detection result 22.
[0122] As mentioned above, because the monitoring system in
accordance with this Embodiment 3 is configured in such a way as to
determine whether or not the number of blocks each of which has a
motion is equal to or greater than the set block number, the
monitoring system can distinguish small moving objects, such as
small animals, and persons.
Embodiment 4
[0123] Although the case in which a picture is partitioned into a
plurality of blocks and the motion determination is performed on
each of the blocks is shown in Embodiments 1 to 3, the blocks after
the partitioning do not have to have a uniform size. In Embodiment
4, a case in which a motion detector 15 (a block partitioner)
partitions a picture (each of inputted and reference pictures) into
blocks having a size in one of regions into which the picture is
divided and blocks having a different size in the other region will
be shown.
[0124] The basic configuration of a monitoring camera 1 in
accordance with Embodiment 4 is the same as that of the monitoring
camera 1 in accordance with Embodiment 1 shown in FIG. 2, and the
explanation will be made by focusing on a different portion.
Although the following explanation will be made by focusing on a
case in which the monitoring camera 1 performs motion
determination, also in the case of Embodiment 4, a picture acquired
by the monitoring camera 1 can be inputted to a network recorder 4
and the network recorder 4 can alternatively perform the motion
determination on the basis of the picture inputted thereto, like in
the case of Embodiment 1. Further, also in the case of Embodiment
4, the picture acquired by the monitoring camera 1 can be
alternatively inputted to a PC for monitoring 3 and the PC for
monitoring 3 can alternatively perform the motion determination on
the basis of the picture inputted thereto, like in the case of
Embodiment 1.
[0125] Further, information indicating the sizes of the blocks in
the two or more regions into which the picture is divided can be
acquired, as a setting parameter 21, from the PC for setting 6 by
an input output unit 13 (a block information acquirer).
[0126] An example of a method of partitioning a picture into blocks
in accordance with Embodiment 4 will be shown in FIG. 13. While a
picture is partitioned into blocks having a uniform size in the
example of FIG. 3, blocks in a region (a zone A) farther from the
monitoring camera 1 are set to have a size smaller than that of
blocks in a region (a zone B) closer to the monitoring camera 1, in
the example of FIG. 13. This is because an object exhibits a size
different dependently upon the distance from the monitoring camera
1 to the target area. By, in this way, partitioning a picture into
blocks while making the sizes of blocks be different in
consideration of the distance between the monitoring camera 1 and
the target area when identifying a moving object within the
picture, the monitoring system can perform the motion determination
corresponding to the distance from the monitoring camera 1 to the
target area on any position within the picture of the target to be
monitored.
[0127] When an object moving far from the monitoring camera 1 and
another object moving close to the monitoring camera 1 are moving
at the same speed, their movement distances in appearance within
the screen differ from each other. Therefore, when a person seen in
the zone A farther from the monitoring camera 1 and another person
seen in the zone B closer to the monitoring camera 1 are moving at
the same speed in the example of FIG. 13, the motion of the person
in the zone A in the screen is observed to be small while the
motion of the other person in the zone B in the screen is observed
to be large.
[0128] In a case in which all the blocks have a uniform size within
the screen, when persons have the same motion within the screen,
the motion of a person moving in a region closer to the monitoring
camera 1 is observed to be large while the motion of another person
moving in a region farther from the monitoring camera 1 is observed
to be small, and therefore the motion of any object can be
determined by taking into consideration the distance from the
monitoring camera 1.
[0129] In contrast with this, by making the sizes of blocks be
different according to the distance from the monitoring camera 1,
objects can be made to move just one block during the same time
period. More specifically, when persons are moving at the same
speed within the screen, the distance which a person can move in a
region farther from the monitoring camera 1 within a time period
during which another person moves between blocks in a region closer
to the monitoring camera 1 is set to one block. As a result,
although the blocks in the zone closer to the monitoring camera 1
have a size larger than those in the other zone farther from the
monitoring camera 1, when persons are moving at the same speed in
both the groups of blocks, they can be moved respectively between
blocks within the same time period or time periods close to each
other.
[0130] By then determining the motion determination index value by
using an individual sensitivity parameter for each of the groups of
blocks closer to and blocks farther from the monitoring camera 1,
the monitoring system can perform the same motion determination as
that on blocks having a uniform size after the partitioning.
[0131] Although the case in which each picture is divided into two
regions: the zone A and the zone B, and the size of blocks in one
of the regions is made to be different from that of blocks in the
other region is shown in Embodiment 4, the sizes of blocks within
the picture can be arbitrary, and a region closer to the monitoring
camera 1 can be partitioned into larger parts and a region farther
from the monitoring camera 1 can be partitioned into finer parts.
The present invention is not limited to those examples.
[0132] As mentioned above, because the monitoring system in
accordance with this Embodiment 4 is configured in such a way as to
make the sizes of blocks in a plurality of regions, into which each
of the pictures (the inputted picture and the reference picture) is
partitioned, be different and perform the block partitioning, the
monitoring system can perform the motion determination
corresponding to the distance from the monitoring camera 1 to the
target area on any position within the picture of the target to be
monitored.
Embodiment 5
[0133] In Embodiments 1 to 4, the case in which the sensitivity
parameter is set, as a setting parameter 21, by the PC for setting
6 according to the distance information of each block (the
information indicating the distance between the target area seen in
the block and the monitoring camera 1) or the optical influence
information (the information indicating whether or not there is an
influence of a light source upon the target area seen in the block,
the degree of the influence, and the type of the influence) is
shown. In contrast with this, the distance information can be set,
as a setting parameter 21, by the PC for setting 6, and this
distance information can be converted into the sensitivity
parameter and this sensitivity parameter can be used. The
conversion of the distance information into the sensitivity
parameter can be performed by the PC for setting 6, or can be
alternatively performed by the monitoring camera 1. Also in the
case of Embodiment 5, a picture acquired by the monitoring camera 1
can be inputted to the network recorder 4 and the network recorder
4 can alternatively perform the motion determination on the basis
of the picture inputted thereto, like in the case of Embodiment 1.
Further, also in the case of Embodiment 5, the picture acquired by
the monitoring camera 1 can be alternatively inputted to the PC for
monitoring 3 and the PC for monitoring 3 can alternatively perform
the motion determination on the basis of the picture inputted
thereto, like in the case of Embodiment 1. In the case in which the
monitoring camera 1 performs the above-mentioned conversion, a
function (a distance acquirer) of acquiring the distance
information is disposed in the input output unit 13, and the
conversion function (a detection sensitivity acquirer) is disposed
in either one of the input output unit 13, the storage unit 14 and
the motion detector 15. By setting the sensitivity parameter
according to the distance information for each block in this way,
erroneous motion detection and omissions in motion detection can be
prevented.
[0134] The distance information can be the absolute value of the
distance or can be index information. For example, the index
information is one indicating "far" or "close, "far", "middle" or
"close, "distance 1", "distance 2", . . . or "distance N", or the
like. As an alternative, as the index information, information
indicating two-dimensional space or three-dimensional space, such
as "horizontal distance 1", . . . or "horizontal distance N", or
"vertical distance 1", . . . or "vertical distance N", can be used.
More specifically, the index information can be arbitrary as long
as the index information is one indicating the distance from the
monitoring camera 1 to the target area, and the present invention
is not limited to those examples.
[0135] Further, the optical influence information can be set, as a
setting parameter 21, by the PC for setting 6, and this optical
influence information can be converted into the sensitivity
parameter and this sensitivity parameter can be used. The
conversion of the optical influence information into the
sensitivity parameter can be performed by the PC for setting 6, or
can be alternatively performed by the monitoring camera 1. In the
case in which the monitoring camera 1 performs the above-mentioned
conversion, a function (an optical influence information acquirer)
of acquiring the optical influence information is disposed in the
input output unit 13, and the conversion function (a detection
sensitivity acquirer) is disposed in either one of the input output
unit 13, the storage unit 14 and the motion detector 15. By setting
the sensitivity parameter according to the optical influence
information for each block in this way, erroneous motion detection
and omissions in motion detection due to the influence of light can
be prevented.
[0136] As a concrete example of the type of the influence of light,
the influence type being indicated by the optical influence
information, there is "a flicker", "fluctuations", "blinking", or
the like. As the optical influence information, all the
above-mentioned pieces of information including "a flicker",
"fluctuations" and "blinking" can be used, or a combination of one
or more pieces of those pieces of information can be used. As an
alternative, the optical influence information can be light
information about direct light, light information about indirect
light caused by reflection, light information about an artificial
light source (a fluorescent lamp, an electric lamp, an LED, or the
like), or other light information (any of the above-mentioned
pieces of light information is either one of "a flicker",
"fluctuations" and "blinking", or a combination of those pieces of
light information), and the present invention is not limited to
those examples.
[0137] Further, the case in which the set block number is set, as a
setting parameter 21, by the PC for setting 6 is shown in
Embodiment 3. In contrast with this, the distance information can
be set, as a setting parameter 21, by the PC for setting 6, and
this distance information can be converted into the set block
number and this set block number can be used. The conversion of the
distance information into the set block number can be performed by
the PC for setting 6, or can be alternatively performed by the
monitoring camera 1. In the case in which the monitoring camera 1
performs the above-mentioned conversion, a function (a distance
acquirer) of acquiring the distance information is disposed in the
input output unit 13, and the conversion function (a set block
number acquirer) is disposed in either one of the input output unit
13, the storage unit 14 and the motion detector 15. By setting the
set block number according to the distance information for each
block in this way, erroneous motion detection and omissions in
motion detection can be prevented.
[0138] As mentioned above, the monitoring system in accordance with
this Embodiment 5 can prevent erroneous motion detection and
omissions in motion detection by setting the distance information
or the optical influence information for each block, and converting
this information into the sensitivity parameter or the set block
number and then using the sensitivity parameter or the set block
number for the motion determination.
Embodiment 6
[0139] In Embodiment 6, a case of combining Embodiments 1 to 5 will
be shown. In an example shown hereafter, a case of performing, as
primary motion determination, the motion determination based on the
sensitivity parameter for each block, which is shown in Embodiment
1, performing, as secondary motion determination, the motion
determination based on the consecutive frame number, which is shown
in Embodiment 2, and performing, as tertiary motion determination,
the motion determination based on the set block number, which is
shown in Embodiment 3, will be shown. Although the following
explanation will be made by focusing on a case in which the
monitoring camera 1 performs the motion determination, also in the
case of Embodiment 6, a picture acquired by the monitoring camera 1
can be inputted to the network recorder 4 and the network recorder
4 can alternatively perform the motion determination on the basis
of the picture inputted thereto, like in the case of Embodiment 1.
Further, also in the case of Embodiment 6, the picture acquired by
the monitoring camera 1 can be alternatively inputted to the PC for
monitoring 3 and the PC for monitoring 3 can alternatively perform
the motion determination on the basis of the picture inputted
thereto, like in the case of Embodiment 1.
[0140] The motion determination process performed by the motion
detector 15 in accordance with Embodiment 6 will be explained by
using FIG. 14. Hereafter, a case in which the sensitivity parameter
for each block is acquired, as a setting parameter 21, from the
storage unit 14 will be shown.
[0141] In the motion determination process by the motion detector
15, the motion detector first, in step ST1401, partitions each of
inputted and reference pictures into a plurality of blocks. As an
alternative, the partitioning of each of the inputted and reference
pictures into a plurality of blocks can be performed by either the
picture processing unit 12 or the storage unit 14. In that case,
the motion detector 15 acquires the inputted picture on which the
block partitioning is performed and the reference picture on which
the block partitioning is performed. The motion detector 15 can be
further configured in such a way as to read data about each block
from either the picture processing unit 12 or the storage unit 14
(addressing).
[0142] The motion detector then, in step ST1402, calculates the
motion determination index value for each block on the basis of the
inputted picture and the reference picture.
[0143] Next, the motion detector, in step ST1403, performs the
primary motion determination. As a concrete example, the motion
detector determines whether or not the motion determination index
value is equal to or greater than the corresponding sensitivity
parameter for each block, like that in accordance with Embodiment
1. When the motion determination index value is equal to or greater
than the corresponding sensitivity parameter, the motion detector
shifts to step ST1404 and determines that "there is a motion
(determination 1)." In contrast, when the motion determination
index value is less than the corresponding sensitivity parameter,
the motion detector shifts to step ST1405 and determines that
"there is no motion (determination 1)."
[0144] Next, the motion detector, in step ST1406, performs the
secondary motion determination. As a concrete example, the motion
detector determines whether or not the number of consecutive frames
each of which is determined, through the primary motion
determination, to be a one in which "there is a motion
(determination 1)" is equal to or greater than the consecutive
frame number, on a per block basis. When the number of consecutive
frames each of which is determined as above is equal to or greater
than the consecutive frame number, the motion detector shifts to
step ST1407 and determines that "there is a motion (determination
2)." In contrast, when the number of consecutive frames each of
which is determined as above is less than the consecutive frame
number, the motion detector shifts to step ST1408 and determines
that "there is no motion (determination 2)."
[0145] Next, the motion detector, in step ST1409, performs the
tertiary motion determination. As a concrete example, the motion
detector determines whether or not the number of blocks each of
which is determined, through the primary motion determination, to
be a one in which "there is a motion (determination 2)" is equal to
or greater than the set block number, like that in accordance with
Embodiment 3. When the number of blocks each of which is determined
as above is equal to or greater than the set block number, the
motion detector shifts to step ST1410 and determines that "there is
a motion (determination 3)." In contrast, when the number of blocks
each of which is determined as above is less than the set block
number, the motion detector shifts to step ST1411 and determines
that "there is no motion (determination 3)." In the tertiary motion
determination, the motion detector can alternatively determine
whether or not the number of blocks adjacent (consecutive) within
the picture each of which is determined to be a one in which "there
is a motion (determination 2)" is equal to or greater than the set
block number.
[0146] The motion detector then, in step ST1412, outputs the motion
detection result 22 acquired through the secondary motion
determination to the input output unit 13. The motion detector can
be alternatively configured in such a way as to, only when the
motion detection result 22 shows "there is a motion (determination
3)", output that motion detection result 22.
Embodiment 7
[0147] In Embodiment 1, the case in which the monitoring system
performs the motion determination by using the sensitivity
parameter set according to the distance from the monitoring camera
1 to the target area is shown. More specifically, the case in which
the reference point for the distance to the target area is the
installation position of the monitoring camera 1 is shown.
[0148] In contrast with this, in Embodiment 7, a case in which the
reference point is not the installation position of the monitoring
camera 1 will be shown. The performance of motion determination
using the sensitivity parameter set according to the distance from
the reference point to a target area is the same as that shown in
Embodiment 1. Although the following explanation will be made by
focusing on a case in which a monitoring camera 1 performs the
motion determination, a picture acquired by the monitoring camera 1
can be inputted to a network recorder 4 and the network recorder 4
can alternatively perform the motion determination on the basis of
the picture inputted thereto, like in the case of Embodiment 1.
Further, also in the case of Embodiment 7, the picture acquired by
the monitoring camera 1 can be alternatively inputted to a PC for
monitoring 3 and the PC for monitoring 3 can alternatively perform
the motion determination on the basis of the picture inputted
thereto, like in the case of Embodiment 1.
[0149] FIG. 15 is a diagram explaining a relation between the
reference point and the distance from the reference point to the
target area. Hereafter, for convenience's sake, the explanation
will be made by assuming that the target area is partitioned into
blocks at equal intervals, but the present invention is not limited
to that example. Further, the explanation will be made by assuming
that the distance to the target area is the distance to a midpoint
of the target area, but the present invention is not limited to
that example.
[0150] FIG. 15(a) is a diagram showing a case in which the
reference point shown in Embodiment 1 is the installation position
of the monitoring camera 1. In this case, the distance from the
reference point to the target area a is the one c (they are 5
blocks apart from each other). Further, the distance from the
reference point to the target area b is the one d (they are 2
blocks apart from each other).
[0151] FIG. 15(b) is a diagram showing a case in which the
reference point is at a position farthest from the installation
position of the monitoring camera 1 (at the farthest position where
the monitoring camera 1 can focus). In this case, the distance from
the reference point to the target area a is the one d' (they are 2
blocks apart from each other). Further, the distance from the
reference point to the target area b is the one c' (they are 5
blocks apart from each other).
[0152] FIG. 15(c) is a diagram showing a case in which the
reference point is at a position between the installation position
of the monitoring camera 1 and the farthest position from the
monitoring camera 1. In this case, the distance from the reference
point to the target area a is the one d (they are 2 blocks apart
from each other). The distance from the reference point to the
target area b is the one d' (they are 2 blocks apart from each
other).
[0153] As a concrete example of a case in which the reference point
is not the installation position of the monitoring camera 1, there
can be a case in which the reference point is a point, a target
area, a block or a zone, within the picture, where the monitoring
camera 1 focuses most clearly (referred to as a just focused
region).
[0154] In conventional technologies, in the case in which the
reference point is a just focused region, the following problem
arises.
[0155] For example, when persons are moving similarly in a region
closer to and in a region farther from the just focused region
within the picture, there may occur a phenomenon in which the
motion of a person moving in a region closer to the just focused
region is easily detected while the motion of another person moving
in a region farther from the just focused region is hard to detect
if the sensitivity parameter is set to a single certain value. This
is because in the case of using a method of performing the motion
determination according to a change state of the pattern of an
object, a change state of the pixel level of a target area to be
monitored, or the like, a fine pattern or the like of a person who
is closer to the just focused region can be observed while a fine
pattern or the like of another person who is farther from the just
focused region collapses and cannot be observed. As a result, there
occurs a phenomenon in which while the motion of a person closer to
the just focused region is easily detected, the motion of another
person farther from the just focused region is hard to detect.
Therefore, there arises a problem that there is a case in which the
motion of a person farther from the just focused region is not
detected.
[0156] In order to solve the above-mentioned problem, there can be
considered a method of increasing the sensitivity for motion
detection (making it possible to determine that "there is a motion"
even though the number of differences is small). However, in this
case, there is an increase in the frequency of erroneous detection
to determine that "there is a motion" even from a small difference
which is not a person's motion.
[0157] Further, in order to solve the above-mentioned problem,
there can be considered a method of lowering the sensitivity for
motion detection (making it possible not to determine that "there
is a motion" even when the signal level changes a little). However,
in this case, there is an increase in omissions in detection
including non-detection of the motion of a person and non-detection
of the motion of a target to be monitored.
[0158] Next, a concrete example of a method of performing the
motion determination by using the sensitivity parameter set
according to the distance from that reference point to the target
area in the case in which the reference point is not the
installation position of the monitoring camera 1 will be shown
below. Hereafter, the explanation will be made by focusing on a
portion different from Embodiment 1.
[0159] In Embodiment 7, sensitivity parameters (first and second
sensitivity parameters) which are different for a region closer to
the just focused region and for a region farther from the just
focused region are set to the picture. As a result, when persons
are moving similarly, not only as to the motion determination index
value acquired for a person moving in a region closer to the just
focused region, but also as to the motion determination index value
acquired for a person moving in a region farther from the just
focused region, the motion detector can determine that "there is a
motion" according to the value.
[0160] As mentioned above, in accordance with Embodiment 7, in
order to perform the motion determination in consideration of the
difference in the detection sensitivity, the difference being
dependent upon the distance from the just focused region to the
target area, the motion detector 15 (the block partitioner)
partitions each picture into a plurality of blocks and the PC for
setting 6 sets the sensitivity parameter for each of the blocks. At
that time, the PC for setting 6 sets the sensitivity parameter
according to the distance between the target area in each of the
blocks and the just focused region. More specifically, because it
is hard to detect the motion of a target staying farther from the
just focused region as compared with that of a target staying
closer to the just focused region, the PC for setting sets the
sensitivity parameter for a region farther from the just focused
region to a value which makes it easier to detect a motion, as
compared with the sensitivity parameter for a region closer to the
just focused region.
[0161] Further, instead of setting the just focused region, the
just focused region can be estimated from the position of the focus
lens which the monitoring camera 1 has, the picture on which the
picture processing is performed by the picture processing unit 12,
or the like.
[0162] The sensitivity parameter is set by the PC for setting 6
according to the distance information of each block (the
information indicating the distance between the target area seen in
the block and the just focused region).
[0163] As an alternative, the above-mentioned distance information
can be set as a setting parameter 21, by the PC for setting 6, and
this distance information can be converted into the sensitivity
parameter and this sensitivity parameter can be used. The
conversion of the distance information into the sensitivity
parameter can be performed by the PC for setting, or can be
alternatively performed by the monitoring camera 1. In the case in
which the monitoring camera 1 performs the above-mentioned
conversion, a function (a distance acquirer) of acquiring the
distance information is disposed in the input output unit 13, and
the conversion function (a detection sensitivity acquirer) is
disposed in either one of the input output unit 13, the storage
unit 14 and the motion detector 15.
[0164] Further, the configuration in accordance with Embodiment 7
can be used while being combined with the configuration in
accordance with Embodiment 1. In that case, what is necessary is
just to perform the motion determination by using the sensitivity
parameter set according to both the distance (1) from the
monitoring camera 1 to the target area, and the distance (2) from
the just focused region to the target area. The sensitivity
parameter can be defined by making the above-mentioned distances
(1) and (2) be variable.
[0165] As shown in FIGS. 15(b) and 15(c), there are two kinds of
the above-mentioned distance (2) from the just focused region (the
reference point) to the target area. More specifically, there are
two cases: a case in which the portion extending from the just
focused region to the target area has a direction going away from
the monitoring camera 1 (the distance d in FIG. 15(c)); and a case
in which the portion extending from the just focused region to the
target area has a direction getting close to the monitoring camera
1 (the distance c' and the distance d' shown in FIG. 15(b), and the
distance d' shown in FIG. 15(c)). Therefore, what is necessary is
just to perform the motion determination by using the sensitivity
parameter set according to the above-mentioned distance (1) from
the monitoring camera 1 to the target area, the distance (2-1) in
the case in which the portion extending from the just focused
region to the target area has a direction going away from the
monitoring camera 1, and the distance (2-2) in the case in which
the portion extending from the just focused region to the target
area has a direction getting close to the monitoring camera 1. The
sensitivity parameter can be defined by making the above-mentioned
distances (1), (2-1) and (2-2) be variable.
Embodiment 8
[0166] In Embodiment 1, the case in which the motion determination
is performed by using the sensitivity parameter set according to
the distance from the monitoring camera 1 to the target area is
shown. In contrast with this, in Embodiment 8, a case in which a
function of changing the zoom magnifying power is disposed in a
monitoring camera 1, and the sensitivity parameter is set (updated)
automatically according to the zoom magnifying power will be shown.
Although the following explanation will be made by focusing on a
case in which the monitoring camera 1 performs motion
determination, also in the case of Embodiment 8, a picture acquired
by the monitoring camera 1 can be inputted to a network recorder 4
and the network recorder 4 can alternatively perform the motion
determination on the basis of the picture inputted thereto, like in
the case of Embodiment 1. Further, also in the case of Embodiment
8, the picture acquired by the monitoring camera 1 can be
alternatively inputted to a PC for monitoring 3 and the PC for
monitoring 3 can alternatively perform the motion determination on
the basis of the picture inputted thereto, like in the case of
Embodiment 1.
[0167] The monitoring camera 1 has a function of changing the zoom
magnifying power by using a zoom lens (zooming in (telephoto) and
zooming out (wide angle)). A detection sensitivity acquirer then
sets (updates) the corresponding sensitivity parameter
automatically according to the zoom magnifying power set by the
monitoring camera 1.
[0168] First, an automatic setting (update) of the sensitivity
parameter at the time when the monitoring camera 1 zooms in will be
explained.
[0169] It is assumed that in an example of FIG. 16(a), the
monitoring camera 1 zooms in and enlarges a region C in a zone A,
the region being far from the monitoring camera 1, and a picture
shown in FIG. 16(b) is acquired. By zooming in the region by using
the monitoring camera 1, as shown in FIG. 16 (b), objects can be
observed to have sizes similar to those of objects at shorter
distances from the monitoring camera 1. More specifically, the
distance from the monitoring camera 1 can be changed in appearance
according to the zoom magnifying power of the monitoring camera 1.
Therefore, the detection sensitivity acquirer updates the
sensitivity parameter independently for each block or each zone
within the screen according to the zoom magnifying power.
[0170] As a concrete example of the method of updating the
sensitivity parameter according to the zoom magnifying power, there
is a method (1) of multiplying the individual sensitivity parameter
for each block or each zone by a threshold conversion parameter
corresponding to the zoom magnifying power and setting the acquired
sensitivity parameter to that block or that zone. Further, there is
a method (2) of referring to a sensitivity parameter table
corresponding to the zoom magnifying power and setting the
sensitivity parameter corresponding to each block or each zone.
[0171] Further, also when the monitoring camera 1 zooms out, the
same process as that as above is performed. When the screen becomes
a wide angle because of the zooming out, objects can be observed to
have sizes similar to those of objects at longer distances from the
monitoring camera 1. More specifically, the distance from the
monitoring camera 1 can be changed in appearance according to the
zoom magnifying power of the monitoring camera 1. Therefore, the
detection sensitivity acquirer updates the sensitivity parameter
independently for each block or each zone within the screen
according to the zoom magnifying power.
[0172] A concrete example of the method of updating the sensitivity
parameter according to the zoom magnifying power is the same as
that at the time of the above-mentioned zooming in, the explanation
of the concrete example will be omitted hereafter.
Embodiment 9
[0173] In Embodiments 1 to 8, the distance between each object and
the monitoring camera 1 can be measured by using a sensor. Although
the following explanation will be made by focusing on a case in
which the monitoring camera 1 performs the motion determination,
also in the case of Embodiment 9, a picture acquired by the
monitoring camera 1 can be inputted to the network recorder 4 and
the network recorder 4 can alternatively perform the motion
determination on the basis of the picture inputted thereto, like in
the case of Embodiment 1. Further, also in the case of Embodiment
9, the picture acquired by the monitoring camera 1 can be
alternatively inputted to the PC for monitoring 3 and the PC for
monitoring 3 can alternatively perform the motion determination on
the basis of the picture inputted thereto, like in the case of
Embodiment 1.
[0174] The sensor is an ultrasonic sensor mounted to the monitoring
camera 1, and measures the distance between each object and the
monitoring camera 1 for each block from the time difference between
the time when an ultrasonic wave is outputted and the time when the
outputted ultrasonic wave reflected by that object returns
thereto.
[0175] The detection sensitivity acquirer sets, on the basis of the
distance for each block which is measured by the sensor, the
sensitivity parameter according to that distance for that block or
each zone within the screen.
[0176] In the measurement on each block, the distance to a pixel at
a predetermined position (e.g., at the center, a corner or another
representative position of the block, or in a specified region of
the block) or to a specified region is measured (completely
automatically). As an alternative, the distance to a portion (a
pixel or a specified region) in each block of the monitor screen,
the portion being specified by a surveillant, is measured
(semi-automatically). As an alternative, the distance to a
predetermined object (the distance to the specified object) with a
region enclosed by edges being set as an object is measured. Any
other method is used as long as it can measure the distance for
each block can be alternatively used, and the present invention is
not limited to those examples.
[0177] Although an ultrasonic sensor is used as the sensor in
Embodiment 9, an infrared sensor or another sensor can be
alternatively used. More specifically, any other sensor is used as
long as it can send a radar, an acoustic wave, a radio wave, or a
signal wave and measure the distance from the time difference
between the time when the radar, the acoustic wave, the radio wave,
or the signal wave is outputted and the time when the radar, the
acoustic wave, the radio wave, or the signal wave reflected by an
object returns thereto, and the present invention is not limited to
those examples.
[0178] In addition, while the invention has been described in its
preferred embodiments, it is to be understood that an arbitrary
combination of two or more of the embodiments can be made, various
changes can be made in an arbitrary component in accordance with
any one of the embodiments, and an arbitrary component in
accordance with any one of the embodiments can be omitted within
the scope of the invention.
INDUSTRIAL APPLICABILITY
[0179] The monitoring camera in accordance with the present
invention can prevent erroneous motion detection and omissions in
motion detection in each of blocks into which a picture is
partitioned, with a simple and low-cost configuration, and is
suitable for use as a monitoring camera or the like that performs
motion determination on the basis of a picture acquired by
performing shooting on a target area.
EXPLANATIONS OF REFERENCE NUMERALS
[0180] 1 monitoring camera, 2 network, 3 PC for surveillance, 4
network recorder, 5 monitor, 6 PC for setting, 10 picture shooter,
11 picture input unit, 12 picture processing unit, 13 input output
unit, 14 storage unit, 15 motion detector, 21 setting parameter, 22
motion detection result, and 31 processed picture.
* * * * *